The bar plot shows active cases in three provinces accross Canada. BC active cases data is not maintained by the government of British Columbia so here it is being approximated by summing all reported cases within the last 14 days.
To get a better idea of overall provincial case situation, active cases are normalized against population. Active cases are divided by a population factor of 100k so the outcome becomes Active Cases per 100 000 people.
Provincal data is a bit misleading, especially in a large country like Canada. Time to get a more granular look at active cases across Canada. Most provinces report their case data by a smaller health region. Let’s replicate the previous plots for the local health regions in Alberta, BC, and Ontario. the dataset has been filtered to show only health regions with over 100 active cases.
Next we look at active COVID cases vs population for local health regions across Alberta, BC and Ontario. As expected, a somewhat linear relationship exists between population and Active COVID cases.
Now let’s look at those plots normalized to local health region populations.
As we’re often reminded, distance between people limits the spread of the virus. In an attempt to account for regions where distance between people is somewhat unavoidable we will apply a penalty to active COVID cases based on population density within local health regions. To keep it simple, Active Cases are multiplied by population density which is a simple measure of population per square kilometer. When shown on a regular scale the disparity between the Toronto/Peel and all other regions becomes apparent.
Alberta daily reported cases smoothed using a loess function to remove local variability.
Ontario daily reported cases smoothed using a loess function to remove local variability. The dataset has been filtered to the 7 regions with the highest active cases.
British Columbia daily reported cases using the provincially reported smothed dataset.